I can’t agree with your theory here. The issue with Net Neutrality is not (as you imply above) the question of whether you pay for a larger pipe to get better performance. It is the question of how many times you pay for the pipe. The internet is mostly comprised of networks that are connected together at peering points (paid and unpaid) where ISPs exchange data as quickly and in as much volume as they can. As long as there is bandwith, all data flows. Generally, when a corporation purchase a 1.544Mbps connection (T-1), they are guaranteed that from their facility to their ISP, they have a dedicated 1.544Mbps in each direction. This is completely paid for and they can send or receive as much data as they like. There is an expectation that their traffic will not be shaped in any way and there is also the expectation that they have paid for both the send and receive directions. This is not to say that you expect you will get 1.544Mbps end-to-end regardless of who you are talking to, that’s going to be dependent upon how well your ISP is connected and is one of the items that is involved in the choice of your ISP.

The “non-neutral” ideas are basically the complete separation of the upstream and downstream connections. Right now, if you pay your cable company for 8Mbps downlink and 300Kbps uplink, you’d expect that you had paid to have up to 8Mbps data coming to your home and 300Kbps going away from your home at any point in time. It is reasonable, since they are selling differentially-priced products, to have this expection, as the difference between 4Mbps and 8Mbps if you can’t have more than 2Mbps into your pipe is pretty much zero.

However, the cable companies and phone companies (who continue to advertise increasing speed and bandwidth so that they can compete with each other) are not interested in paying the price that the commercial ISPs pay to get this additional bandwidth into their systems. Instead, they want to make it the responsibility of the sender (who has already paid their ISP for the outgoing transit of their bits) to pay for those bits to enter their (the cable/telephone company’s) network.

This is where things are problematic. What is the cable company selling you if they are selling you an 8Mbps downlink connection, but that connection doesn’t include the content you want to reach? What are you paying for? How many times do you need to pay for it?

One more thing: QoS has been “around” for years, but it is minimally used because most people who have tried to effectively implement a network based on QoS have quickly found that adding bandwidth is both cheaper and more effective than trying to shape a network with QoS. Ask any major ISP and you’ll find this approach in their backbone.

As to whether you want to pay for higher or lower latency, that’s not what Network Neutrality is about. This isn’t the communism of the net that we’re looking at here, we’re really looking at fairness in advertising and receiving what you pay for. If the phone company wants to sell a 128kbps high-latency “basic” service, nobody’s suggesting they shouldn’t.

However, if the phone company is going to sell 100,000 homes worth of 8Mbps service for a premium over what those homes would pay for a 128Kbps (or 4Mbps) service, then they should be provisioning their network in a way commensurate with the needs of those customers. If they are charging too little, they should charge more and customers will migrate to the 4Mbps connection that the phone company can actually support.

If the phone companies are paid by the consumer for 8Mbps download and then expect the content providers (without whom there would be no interest in an 1Mbps, much less an 8Mbps connection) to pay to put their bits in front of the customers in a free-and-fair manner which is commensurate with what they have paid their network provider for and what the network will handle, then that’s a problem, and that’s where people start complaining about network neutrality.

I think the argument is also about the big sites paying to guarantee their service, if what I’ve read/seen is correct, Google for example. It’s not just the users end that could be ‘at risk’, if such an event were to occur.

What about QOS on a home network?

Your question is a little open-ended, but QOS starts with the OS. XP and Vista both prioritize network use to make sure things like ftp/http don’t interrupt RTSP and other streaming protocols. I do not know enough about home routers to know if they can be setup to enforce QOS, but I will certainly be looking into it if I start streaming videos to my XBox. I really wouldn’t want torrents to make a video stutter.

To be clear, I am not against consumer protection legislation to protect us from greedy ISPs. I just was trying to point out that nothing exists today that you could call ‘Network Neutrality’; so it cannot be comming to an end.

Gaige B Paulsen said: … a lot.. :)

I don’t disagree with anything specific that you said, but none of the “discussions” I have seen on Slashdot, Digg, or similar sites, ever bother to talk about the real issues. They tend to just be a bunch of people saying “they are going to ruin the net”.

One point that I would like to clarify is that when you pay for your home connection, the bandwidth numbers they quote you, only apply the “last mile”. That is, you are paying for specific bandwidth from your house to your ISP’s network. I have never seen a contract that attempted to guarantee a specific amount of bandwidth to the Internet in general. With the exception of some poorly run cable systems, this piece of the network is rarely the problem, and yet it is the only piece that is actually covered in the contract you have with your ISP.

It sounds like we agree that what is needed is legislation (and enforcement of existing regulations) to make sure that everyone understands what they are paying for has the right to get service from someone else if their needs are not being met.